TestMu AI (formerly LambdaTest) MCP server
io.github.LambdaTest/mcpLambdaTeststreamable-httpv1.0.7Scanned
Agentic AI quality engineering: run, debug and triage tests on real browsers and devices
88 tools: 19 can delete or cancel things, 18 change things and 51 only read.
- Reads private data · 7
- Reads text others write · 16
- Can send data out · 5
- 7 findings
Tools listed without sign-in. Remote posture
With Anlyon
19 tools here can delete or cancel things. Govern them.
- Anlyon lists the tools and adopts none.
- Each tool you adopt gets approvals and receipts.
- Your agent never holds the key.
A tool your agent calls directly is not routed through Anlyon and is not governed by it. Anlyon Radar only lists and labels, and these labels are the same either way.
What the TestMu AI (formerly LambdaTest) MCP server can do
The TestMu AI (formerly LambdaTest) MCP server is published by the GitHub account LambdaTest. It has 88 tools. It can delete or cancel things (19 tools: updateBuild, stopBuild, deleteBuild, stopTunnel and 15 more) and create or change data (18 tools: upload_app, generateHyperExecuteYAML, answerHyperExecuteQuery, getAccessibilityReport and 14 more). 51 tools only read.
7 tools read private data (tm_get_test_cases, tm_list_folders, tm_list_modules and 4 more); 16 tools bring in text other people can write (getAccessibilityReport, buildLocalAppForAnalysis, getAutomationTestBrowserConsoleLogs and 13 more); 5 tools can send data out of your system (getAccessibilityReport, postSessionExceptions, postTestExceptions and 2 more). On its own it has all three parts of a prompt-injection exfiltration path: private data, text others write, and a way out.
The fixed checks matched 7 findings (annotation-conflict and credentials-mention). Each one comes with the text it matched.
Its remote endpoint lists its tools without sign-in, and its tools changed once since Anlyon Radar first saw them.
Last scanned Nov 16, 2026, version 1.0.7.
Tools
Grouped by what each tool does, most consequential first. Open one for its reasons and full description; descriptions are the publisher's own words.
88 tools
Deletes or cancels 19Claude Code: deny
deleteBuildPermanently deletes a TestMu AI web automation build and its test results.No findings
Why it is in this group
- The name "deleteBuild" says "delete": a word for removing or ending something.
- The description says "Permanently deletes".
- The server marks it destructive (destructiveHint: true).
Description, as the server wrote it Permanently deletes a TestMu AI web automation build and its test results. Cannot be undone.
Annotations the server declares
- title: Delete Automation Build
- readOnlyHint: false
- destructiveHint: true
- idempotentHint: true
- openWorldHint: false
deletePrerunFileDelete an uploaded prerun file by stored TestMu AI file path.No findings
Why it is in this group
- The name "deletePrerunFile" says "delete": a word for removing or ending something.
- The server marks it destructive (destructiveHint: true).
Description, as the server wrote it Delete an uploaded prerun file by stored TestMu AI file path.
Annotations the server declares
- title: Delete Prerun File
- readOnlyHint: false
- destructiveHint: true
- openWorldHint: false
deleteSessionPermanently delete a specific automation test session from TestMu AI by session ID.No findings
Why it is in this group
- The name "deleteSession" says "delete": a word for removing or ending something.
- The description says "Permanently delete".
- The server marks it destructive (destructiveHint: true).
Description, as the server wrote it Permanently delete a specific automation test session from TestMu AI by session ID.
Annotations the server declares
- title: Delete Session
- readOnlyHint: false
- destructiveHint: true
- openWorldHint: false
deleteUploadedAppsDelete one or more uploaded apps from TestMu AI by their app IDs.No findings
Why it is in this group
- The name "deleteUploadedApps" says "delete": a word for removing or ending something.
- The server marks it destructive (destructiveHint: true).
Description, as the server wrote it Delete one or more uploaded apps from TestMu AI by their app IDs.
Annotations the server declares
- title: Delete Uploaded Apps
- readOnlyHint: false
- destructiveHint: true
- openWorldHint: false
deleteUserFileDelete an uploaded user file by stored TestMu AI file path.No findings
Why it is in this group
- The name "deleteUserFile" says "delete": a word for removing or ending something.
- The server marks it destructive (destructiveHint: true).
Description, as the server wrote it Delete an uploaded user file by stored TestMu AI file path.
Annotations the server declares
- title: Delete User File
- readOnlyHint: false
- destructiveHint: true
- openWorldHint: false
smartui_baseline_manageMerge SmartUI baselines across branches or builds (to read baseline settings, use smartui_project_query with action=get).No findings
Why it is in this group
- The description's first verb is "merge": a word for a change that another call does not undo.
- The server marks it destructive (destructiveHint: true).
Description, as the server wrote it Merge SmartUI baselines across branches or builds (to read baseline settings, use smartui_project_query with action=get). merge_branch/merge_build resolve source and target names to builds, merge them, then poll the merged build and report its real terminal status — a merge that lands in error or with zero screenshots is returned as a tool error, not as success. Merging permanently changes baseline state: the merged build takes the baseline flag, so a failed merge leaves the baseline on an empty build.
Annotations the server declares
- title: SmartUI Baseline Manage
- readOnlyHint: false
- destructiveHint: true
- openWorldHint: false
smartui_screenshot_reviewApprove or reject SmartUI comparisons — individually (comparison_ids or ss_ids), or for a whole build (all_matching:{status}).Reads text others write1 finding
Why it is in this group
- The name "smartui_screenshot_review" says "screenshot": a word for reading.
- The description's first verb is "approve": a word for a change.
- The server marks it destructive (destructiveHint: true).
What it can reach
- untrusted-inputThe name "smartui_screenshot_review" says "screenshot" and "review": it reads comments, which other people can write.
Findings
- Warn
The server marks "smartui_screenshot_review" destructive (destructiveHint: true), and its name says "screenshot", a word for reading.
What this check looks forMatched text
annotations: {"readOnlyHint":false,"destructiveHint":true,"openWorldHint":false}; The name "smartui_screenshot_review" says "screenshot": a word for reading.
Description, as the server wrote it Approve or reject SmartUI comparisons — individually (comparison_ids or ss_ids), or for a whole build (all_matching:{status}). Ids are validated against the build before writing, and results are verified by re-reading status, so the response reports real per-item updated[]/failed[] and whether the baseline was promoted. Approving comparisons on the baseline branch can promote the build to baseline.
Annotations the server declares
- title: SmartUI Screenshot Review
- readOnlyHint: false
- destructiveHint: true
- openWorldHint: false
smartui_screenshot_uploadUpload screenshots to a SmartUI project as a new build, from local file_paths or from source_urls (fetched server-side, so it works without a client filesystem).Reads private dataReads text others write1 finding
Why it is in this group
- The name "smartui_screenshot_upload" says "screenshot": a word for reading.
- The description's first verb is "upload": a word for a change.
- The server marks it destructive (destructiveHint: true).
What it can reach
- private-dataThe description's first sentence says "file": it reads files.
- private-dataThe input names a file (input.file_paths), and the tool reads: it reads files.
- untrusted-inputThe description's first sentence says "upload" and "url": it reads web pages, which anyone can write.
Findings
- Warn
The server marks "smartui_screenshot_upload" destructive (destructiveHint: true), and its name says "screenshot", a word for reading.
What this check looks forMatched text
annotations: {"readOnlyHint":false,"destructiveHint":true,"openWorldHint":false}; The name "smartui_screenshot_upload" says "screenshot": a word for reading.
Description, as the server wrote it Upload screenshots to a SmartUI project as a new build, from local file_paths or from source_urls (fetched server-side, so it works without a client filesystem). Pass project_id to auto-resolve the project token. Set mark_baseline=true to seed a baseline. Returns build_id; follow with smartui_build_query(action=gate).
Annotations the server declares
- title: SmartUI Screenshot Upload
- readOnlyHint: false
- destructiveHint: true
- openWorldHint: false
stopBuildStops every running test in a TestMu AI web automation build.No findings
Why it is in this group
- The name "stopBuild" says "stop": a word for a change.
- The server marks it destructive (destructiveHint: true).
Description, as the server wrote it Stops every running test in a TestMu AI web automation build. Running sessions end immediately.
Annotations the server declares
- title: Stop Automation Build
- readOnlyHint: false
- destructiveHint: true
- idempotentHint: true
- openWorldHint: false
stopSessionStop an active running test session on TestMu AI.No findings
Why it is in this group
- The name "stopSession" says "stop": a word for a change.
- The server marks it destructive (destructiveHint: true).
Description, as the server wrote it Stop an active running test session on TestMu AI.
Annotations the server declares
- title: Stop Session
- readOnlyHint: false
- destructiveHint: true
- openWorldHint: false
stopTunnelTerminate a specific active tunnel connection by tunnel ID.No findings
Why it is in this group
- The name "stopTunnel" says "stop": a word for a change.
- The description's first verb is "terminate": a word for removing or ending something.
- The server marks it destructive (destructiveHint: true).
Description, as the server wrote it Terminate a specific active tunnel connection by tunnel ID. Use this to clean up tunnels that are no longer needed.
Annotations the server declares
- title: Stop Tunnel
- readOnlyHint: false
- destructiveHint: true
- openWorldHint: false
tm_record_test_resultsRecord execution results on a MANUAL run's TEST RUN INSTANCES (the per-run copies of each test case) — max 500 per request, all-or-nothing (larger requests are rejected with nothin…No findings
Why it is in this group
- Nothing in its name or description says what it does, so it is counted as a write.
- The server marks it destructive (destructiveHint: true).
Description, as the server wrote it Record execution results on a MANUAL run's TEST RUN INSTANCES (the per-run copies of each test case) — max 500 per request, all-or-nothing (larger requests are rejected with nothing recorded). Manual runs only: KaneAI run results are recorded by KaneAI executions, so manual entry on a KaneAI run is rejected. Target an instance by instance_id, or just pass test_case_id and the instance is resolved automatically. Valid statuses: Passed, Failed, Skipped, Not Started (terminal — run-level "In Progress" is set via tm_update_test_run). Re-recording a step status replaces its result: an omitted step remark is cleared; response echoes per-step outcomes. Optional remarks and assignee per result, and per-step results via steps[] (mark individual test steps / BDD scenario rows Passed/Failed/Skipped with remarks, like the UI's per-step Mark Status). Note: this sets INSTANCE statuses; the run's overall status is set separately via tm_update_test_run.
Annotations the server declares
- title: Record Test Results (Bulk)
- readOnlyHint: false
- destructiveHint: true
- openWorldHint: false
tm_update_folderRename a folder, change its description, or move it under a different parent folder.No findings
Why it is in this group
- The name "tm_update_folder" says "update": a word for a change.
- The server marks it destructive (destructiveHint: true).
Description, as the server wrote it Rename a folder, change its description, or move it under a different parent folder. Works on all three folder types: per-project Test Cases (default) and Test Runs trees, and the organization-level Modules tree. Moves of test-cases/test-runs folders require project_id; module folders need none.
Annotations the server declares
- title: Update Folder
- readOnlyHint: false
- destructiveHint: true
- openWorldHint: false
tm_update_milestoneUpdate a milestone — title, description, dates, tags — or change its status (e.g.No findings
Why it is in this group
- The name "tm_update_milestone" says "update": a word for a change.
- The server marks it destructive (destructiveHint: true).
Description, as the server wrote it Update a milestone — title, description, dates, tags — or change its status (e.g. mark it Completed). Deleting milestones is not available through this connection.
Annotations the server declares
- title: Update Milestone
- readOnlyHint: false
- destructiveHint: true
- openWorldHint: false
tm_update_projectUpdate a project's name, description or tags (fields are merged — only what you pass changes).No findings
Why it is in this group
- The name "tm_update_project" says "update": a word for a change.
- The server marks it destructive (destructiveHint: true).
Description, as the server wrote it Update a project's name, description or tags (fields are merged — only what you pass changes). Project & Org Instructions stay managed in the UI.
Annotations the server declares
- title: Update Project
- readOnlyHint: false
- destructiveHint: true
- openWorldHint: false
tm_update_test_caseEdit an existing test case (fields are merged — only what you pass changes).No findings
Why it is in this group
- The name "tm_update_test_case" says "update": a word for a change.
- The server marks it destructive (destructiveHint: true).
Description, as the server wrote it Edit an existing test case (fields are merged — only what you pass changes). test_steps, when given, REPLACE the case's steps; module_ids append module reference steps. Deleting test cases is not available through this connection.
Annotations the server declares
- title: Update Test Case
- readOnlyHint: false
- destructiveHint: true
- openWorldHint: false
tm_update_test_runUpdate an open test run — overall status (Passed / Failed / Skipped / In Progress), title, objective, milestones, folder, add test cases (each added case becomes a new test run ins…No findings
Why it is in this group
- The name "tm_update_test_run" says "update": a word for a change.
- The server marks it destructive (destructiveHint: true).
Description, as the server wrote it Update an open test run — overall status (Passed / Failed / Skipped / In Progress), title, objective, milestones, folder, add test cases (each added case becomes a new test run instance; cases must match the run's type — KaneAI or manual), apply a Configuration to the run's instances (configuration_id; on KaneAI runs it must be kane-supported and platform-compatible with the targeted cases), toggle sequential execution (KaneAI runs), or archive it. Only open (active) runs can be updated; a closed/archived run returns a clear message. Adding cases/milestones appends, never replaces.
Annotations the server declares
- title: Update Test Run
- readOnlyHint: false
- destructiveHint: true
- openWorldHint: false
updateBuildRenames a TestMu AI web automation build.No findings
Why it is in this group
- The name "updateBuild" says "update": a word for a change.
- The server marks it destructive (destructiveHint: true).
Description, as the server wrote it Renames a TestMu AI web automation build. Overwrites the current build name.
Annotations the server declares
- title: Rename Automation Build
- readOnlyHint: false
- destructiveHint: true
- idempotentHint: true
- openWorldHint: false
updateSessionUpdate the name or pass/fail status of a test session.No findings
Why it is in this group
- The name "updateSession" says "update": a word for a change.
- The server marks it destructive (destructiveHint: true).
Description, as the server wrote it Update the name or pass/fail status of a test session. Use status_ind to mark tests as passed, failed, error, or skipped from your test framework.
Annotations the server declares
- title: Update Session
- readOnlyHint: false
- destructiveHint: true
- openWorldHint: false
Changes things 18Claude Code: ask
analyzeAppViaTunnelAnalyzes a locally running app for accessibility via TestMu AI tunnel.1 finding
Why it is in this group
- The name "analyzeAppViaTunnel" says "analyze": a word for reading.
- The server marks it as not read-only (readOnlyHint: false).
Findings
- Warn
The server marks "analyzeAppViaTunnel" as not read-only (readOnlyHint: false), and its words say it reads.
What this check looks forMatched text
annotations: {"readOnlyHint":false,"destructiveHint":false,"openWorldHint":false}; The name "analyzeAppViaTunnel" says "analyze": a word for reading.
Description, as the server wrote it Analyzes a locally running app for accessibility via TestMu AI tunnel.
Annotations the server declares
- title: Analyze App via Tunnel
- readOnlyHint: false
- destructiveHint: false
- openWorldHint: false
answerHyperExecuteQueryAnswers questions about HyperExecute features, configuration, usage, troubleshooting, and best practices by searching through documentation.1 finding
Why it is in this group
- The name "answerHyperExecuteQuery" says "execute": a word for a change.
- The server marks it read-only (readOnlyHint: true), but an annotation never lowers a class.
Findings
- Warn
The server marks "answerHyperExecuteQuery" read-only (readOnlyHint: true), and its own name, description or input give the class "write".
What this check looks forMatched text
annotations: {"readOnlyHint":true,"destructiveHint":false,"openWorldHint":false}; The name "answerHyperExecuteQuery" says "execute": a word for a change.
Description, as the server wrote it Answers questions about HyperExecute features, configuration, usage, troubleshooting, and best practices by searching through documentation.
Annotations the server declares
- title: Answer HyperExecute Question
- readOnlyHint: true
- destructiveHint: false
- openWorldHint: false
checkAppProcessingStatusCheck whether an uploaded app has been processed and is ready for features like image injection, network logs, and screenshot unblock on real devices.1 finding
Why it is in this group
- The name "checkAppProcessingStatus" says "status": a word for reading.
- The server marks it as not read-only (readOnlyHint: false).
Findings
- Warn
The server marks "checkAppProcessingStatus" as not read-only (readOnlyHint: false), and its words say it reads.
What this check looks forMatched text
annotations: {"readOnlyHint":false,"destructiveHint":false,"idempotentHint":true,"openWorldHint":false}; The name "checkAppProcessingStatus" says "status": a word for reading.
Description, as the server wrote it Check whether an uploaded app has been processed and is ready for features like image injection, network logs, and screenshot unblock on real devices. The TestMu AI API has no read-only status call: this one starts or refreshes the processing for the requested features, so the app's last-updated time changes. An app id that is not among your uploaded apps returns an error.
Annotations the server declares
- title: Check App Processing Status (Real Device)
- readOnlyHint: false
- destructiveHint: false
- idempotentHint: true
- openWorldHint: false
generateHyperExecuteYAMLGenerates a HyperExecute YAML configuration for running tests on TestMu AI's HyperExecute platform.1 finding
Why it is in this group
- The name "generateHyperExecuteYAML" says "generate": a word for a change.
- The server marks it read-only (readOnlyHint: true), but an annotation never lowers a class.
Findings
- Warn
The server marks "generateHyperExecuteYAML" read-only (readOnlyHint: true), and its own name, description or input give the class "write".
What this check looks forMatched text
annotations: {"readOnlyHint":true,"destructiveHint":false,"openWorldHint":false}; The name "generateHyperExecuteYAML" says "generate": a word for a change.
Description, as the server wrote it Generates a HyperExecute YAML configuration for running tests on TestMu AI's HyperExecute platform. REQUIRED fields: project (language, framework) and testing (testDiscovery.command, testRunnerCommand) All other fields are OPTIONAL. Supports multiple languages and frameworks: - JavaScript/TypeScript: Playwright, Cypress, WebDriverIO, Puppeteer, Taiko - Python: Pytest, Behave, Robot Framework - Java: TestNG, JUnit, Cucumber - Ruby: RSpec, Capybara - C#: NUnit, xUnit, MSTest - Mobile (ONLY use appTesting for these): XCUI (iOS), Espresso (Android), Appium Features: - Smart framework defaults for common configurations - Input validation with security checks - Caching and artifact upload configuration - Git integration via sourcePayload - Mobile app testing support (via optional appTesting field) The appTesting field applies to mobile app testing only; web automation (Selenium, Playwright, Cypress, etc.) leaves it out.
Annotations the server declares
- title: Generate HyperExecute YAML
- readOnlyHint: true
- destructiveHint: false
- openWorldHint: false
getAccessibilityReportRuns an accessibility scan on the given URLs and returns the issues found.Reads text others writeCan send data outNo findings
Why it is in this group
- The name "getAccessibilityReport" says "get": a word for reading.
- The description's first verb is "run": a word for a change.
What it can reach
- untrusted-inputThe server marks it openWorldHint: true: it reaches things outside the server's own data.
- untrusted-inputThe description's first sentence says "run" and "issue": it reads issues, which other people can write.
- untrusted-inputThe input takes a URL (input.urls), and the tool reads it: it reads web pages, which anyone can write.
- outboundThe input takes a URL (input.urls): a request to a URL can carry data to whoever runs it.
Description, as the server wrote it Runs an accessibility scan on the given URLs and returns the issues found. Pass urls to start a scan, or testId to re-read a scan that already ran; exactly one of the two. Results arrive in issueDetails: allIssues (most-severe-first, one page at a time), rules (name and reference link per rule id, listed once), pages (every scanned page with its score and issue count), failedRules (per-rule totals for the whole scan) and pagination. Each issue carries its own howToFix and criteria, measured for that element - two issues sharing a rule can have different causes and fixes. While pagination.nextOffset is not null the results are incomplete: call again with the testId from the response plus that offset, keeping the same filters, until it is null. Page with testId, never by passing urls again - that starts a new scan, costs another run, and can return a different set of issues whose offsets do not line up. a11yReportLink is the dashboard URL and needs a login; shareableLink is public and is only returned by the original scan.
Annotations the server declares
- title: Get Accessibility Report
- readOnlyHint: false
- destructiveHint: false
- openWorldHint: true
postSessionExceptionsUpload exception/error logs from the client machine for a test session, by session ID.Can send data outNo findings
Why it is in this group
- The name "postSessionExceptions" says "post": a word for a change.
What it can reach
- outboundThe name "postSessionExceptions" says "post": it sends or publishes something to other people.
Description, as the server wrote it Upload exception/error logs from the client machine for a test session, by session ID. Use this to attach assertion failures or unexpected errors to a session. The result reads the test's exceptions back to confirm they were stored. The automation dashboard may not show exceptions uploaded this way after the test ended.
Annotations the server declares
- title: Post Session Exceptions
- readOnlyHint: false
- destructiveHint: false
- openWorldHint: false
postTestExceptionsUpload exception logs for a specific test (by test ID) rather than by session ID.Can send data outNo findings
Why it is in this group
- The name "postTestExceptions" says "post": a word for a change.
What it can reach
- outboundThe name "postTestExceptions" says "post": it sends or publishes something to other people.
Description, as the server wrote it Upload exception logs for a specific test (by test ID) rather than by session ID. The result reads the test's exceptions back to confirm they were stored. The automation dashboard may not show exceptions uploaded this way after the test ended.
Annotations the server declares
- title: Post Test Exceptions
- readOnlyHint: false
- destructiveHint: false
- openWorldHint: false
tm_create_folderCreate a folder — in a project's Test Cases tree (default) or Test Runs tree (pass project_id), or in the organization-level Modules tree (no project_id needed; module folders are …No findings
Why it is in this group
- The name "tm_create_folder" says "create": a word for a change.
Description, as the server wrote it Create a folder — in a project's Test Cases tree (default) or Test Runs tree (pass project_id), or in the organization-level Modules tree (no project_id needed; module folders are shared across projects). Optionally nest it under an existing parent folder to build folder trees. Deleting folders is not available through this connection.
Annotations the server declares
- title: Create Folder
- readOnlyHint: false
- destructiveHint: false
- openWorldHint: false
tm_create_milestoneCreate a milestone in a project (status: Open), optionally with description, start/end dates, tags, and existing runs attached.No findings
Why it is in this group
- The name "tm_create_milestone" says "create": a word for a change.
Description, as the server wrote it Create a milestone in a project (status: Open), optionally with description, start/end dates, tags, and existing runs attached.
Annotations the server declares
- title: Create Milestone
- readOnlyHint: false
- destructiveHint: false
- openWorldHint: false
tm_create_projectCreate a new Test Manager project.No findings
Why it is in this group
- The name "tm_create_project" says "create": a word for a change.
Description, as the server wrote it Create a new Test Manager project. Project & Org Instructions (Memory Layer) stay managed in the UI. Deleting projects is not available through this connection.
Annotations the server declares
- title: Create Project
- readOnlyHint: false
- destructiveHint: false
- openWorldHint: false
tm_create_test_caseCreate a single test case in a project folder, with optional steps and references to existing step Modules (module steps are included; the Module itself is unchanged).No findings
Why it is in this group
- The name "tm_create_test_case" says "create": a word for a change.
Description, as the server wrote it Create a single test case in a project folder, with optional steps and references to existing step Modules (module steps are included; the Module itself is unchanged). For creating many cases from a requirement, prefer tm_generate_test_cases.
Annotations the server declares
- title: Create Test Case
- readOnlyHint: false
- destructiveHint: false
- openWorldHint: false
tm_create_test_runCreate a test run from chosen test cases — manual (default) or KaneAI (run_type: "kaneai").No findings
Why it is in this group
- The name "tm_create_test_run" says "create": a word for a change.
Description, as the server wrote it Create a test run from chosen test cases — manual (default) or KaneAI (run_type: "kaneai"). Each included test case becomes a TEST RUN INSTANCE — the executable copy of that test inside the run, which carries its own status, remarks and exactly one Configuration. The cases must match the run type: KaneAI runs only take KaneAI-authored cases with completed code generation; manual runs only take manual cases. Optionally attach existing milestones, file into a run folder, and apply a Configuration to all instances (KaneAI configurations must be compatible with each case's platform — web/mobile-app/mobile-browser and iOS/Android).
Annotations the server declares
- title: Create Test Run
- readOnlyHint: false
- destructiveHint: false
- openWorldHint: false
tm_generate_test_casesGenerate structured test cases from a requirement or prompt using the Test Manager AI generator (applies the project's Memory Layer — Project & Org Instructions — automatically) an…No findings
Why it is in this group
- The name "tm_generate_test_cases" says "generate": a word for a change.
Description, as the server wrote it Generate structured test cases from a requirement or prompt using the Test Manager AI generator (applies the project's Memory Layer — Project & Org Instructions — automatically) and file them into the chosen folder. Requirement documents, specs, spreadsheets or screenshots can be attached via files to ground the generation. Volume = test_scenario_limit × per_scenario_test_cases_limit, capped at 50 per call. By default all generated cases are saved into the folder (one sub-folder per scenario); pass auto_save=false to review before saving. The call returns FAST with a request_id (and a browser progress URL) — generation takes ~30-90s in the background; call this tool again with just request_id to check progress, fetch results, or save. Note: consumes AI generation credits.
Annotations the server declares
- title: Generate Test Cases with AI
- readOnlyHint: false
- destructiveHint: false
- openWorldHint: false
tm_link_ado_issueLink a test case, test run, or a run result (instance) to an Azure DevOps work item (requirement / user story / bug) by its URL — linked test cases are what give the work item its …Can send data outNo findings
Why it is in this group
- The name "tm_link_ado_issue" says "link": a word for a change.
What it can reach
- outboundThe name "tm_link_ado_issue" says "link" and "issue": it writes something other people read.
Description, as the server wrote it Link a test case, test run, or a run result (instance) to an Azure DevOps work item (requirement / user story / bug) by its URL — linked test cases are what give the work item its test coverage and traceability. Requires the Azure DevOps integration to be set up for the organization — otherwise a clear "set up Azure DevOps first" message is returned.
Annotations the server declares
- title: Link Azure DevOps Work Item
- readOnlyHint: false
- destructiveHint: false
- openWorldHint: false
tm_link_jira_issueLink a test case, test run, or a run result (instance) to a Jira issue (requirement / user story / epic / bug) — linked test cases are what give the issue its test coverage and tra…Can send data outNo findings
Why it is in this group
- The name "tm_link_jira_issue" says "link": a word for a change.
What it can reach
- outboundThe name "tm_link_jira_issue" says "link" and "issue": it writes something other people read.
Description, as the server wrote it Link a test case, test run, or a run result (instance) to a Jira issue (requirement / user story / epic / bug) — linked test cases are what give the issue its test coverage and traceability. Linking a result also links its case and run. Requires the Jira integration to be set up for the organization — otherwise a clear "set up Jira first" message is returned.
Annotations the server declares
- title: Link Jira Issue
- readOnlyHint: false
- destructiveHint: false
- openWorldHint: false
triggerSuiteRunTriggers a phone call test run for a specific suite.Reads text others writeNo findings
Why it is in this group
- The name "triggerSuiteRun" says "trigger": a word for a change.
What it can reach
- untrusted-inputThe server marks it openWorldHint: true: it reaches things outside the server's own data.
Description, as the server wrote it Triggers a phone call test run for a specific suite. This will initiate phone calls for all scenarios configured in the suite.
Annotations the server declares
- title: Trigger Suite Run
- readOnlyHint: false
- destructiveHint: false
- openWorldHint: true
upload_appUploads an app to TestMu AI for mobile testing and displays upload status.No findings
Why it is in this group
- The name "upload_app" says "upload": a word for a change.
Description, as the server wrote it Uploads an app to TestMu AI for mobile testing and displays upload status.
Annotations the server declares
- title: Upload App to LambdaTest
- readOnlyHint: false
- destructiveHint: false
- openWorldHint: false
uploadAppVirtualDeviceUpload an app (APK for emulator, IPA for simulator) to TestMu AI for virtual device (emulator/simulator) testing.No findings
Why it is in this group
- The name "uploadAppVirtualDevice" says "upload": a word for a change.
Description, as the server wrote it Upload an app (APK for emulator, IPA for simulator) to TestMu AI for virtual device (emulator/simulator) testing.
Annotations the server declares
- title: Upload App for Virtual Device
- readOnlyHint: false
- destructiveHint: false
- openWorldHint: false
Only reads 51Show 51 read-only tools
analyzeHyperExecuteJobPerform a full analysis of a HyperExecute job: fetches job info, sessions, traverses all tasks → scenario stages → console logs, and returns a structured report with test results, …No findings
Why it is in this group
- The name "analyzeHyperExecuteJob" says "analyze": a word for reading.
Description, as the server wrote it Perform a full analysis of a HyperExecute job: fetches job info, sessions, traverses all tasks → scenario stages → console logs, and returns a structured report with test results, per-scenario logs, and pass/fail statistics. Use this for questions like "which tests failed and why" or "show me the console logs for job X".
Annotations the server declares
- title: Analyze HyperExecute Job
- readOnlyHint: true
- destructiveHint: false
- openWorldHint: false
automationQueryRead-only query of TestMu AI automation resources: list builds, get one build, list the sessions in a build, fetch test details, and retrieve test logs (command, network, console, …No findings
Why it is in this group
- The name "automationQuery" says "query": a word for reading.
Description, as the server wrote it Read-only query of TestMu AI automation resources: list builds, get one build, list the sessions in a build, fetch test details, and retrieve test logs (command, network, console, selenium, instrumentation). Every combination works on platform "web" or "app": build list or get, session list or get, test get, and logs get with a logType. List results carry a pagination object; trust it, not result_set.total, which also counts older builds the list no longer returns. logType "instrumentation" returns XCUI and Espresso test logs. Renaming, stopping and deleting builds are separate tools: updateBuild, stopBuild and deleteBuild.
Annotations the server declares
- title: Automation Query
- readOnlyHint: true
- destructiveHint: false
- openWorldHint: false
buildLocalAppForAnalysisReturns step-by-step instructions to build and serve a local web app and open a TestMu AI tunnel to it, so its accessibility can then be scanned with analyzeAppViaTunnel.Reads text others writeNo findings
Why it is in this group
- The description's first verb is "return": a word for reading.
What it can reach
- untrusted-inputThe description's first sentence says "return" and "web": it reads web pages, which anyone can write.
Description, as the server wrote it Returns step-by-step instructions to build and serve a local web app and open a TestMu AI tunnel to it, so its accessibility can then be scanned with analyzeAppViaTunnel. It does not build, run or scan anything itself, and never includes the access key.
Annotations the server declares
- title: Build and Analyze App
- readOnlyHint: true
- destructiveHint: false
- openWorldHint: false
downloadJobArtifactGet an authenticated download URL (plus size/status metadata) for a specific HyperExecute job artifact.Reads text others write1 finding
Why it is in this group
- The description's first verb is "get": a word for reading.
What it can reach
- untrusted-inputThe description's first sentence says "get" and "url": it reads web pages, which anyone can write.
Findings
- Info
The description mentions "credentials".
What this check looks forMatched text
description: "...urns a URL to fetch with your TestMu AI credentials rather than the raw file bytes."
Description, as the server wrote it Get an authenticated download URL (plus size/status metadata) for a specific HyperExecute job artifact. Requires artefactName (from getHyperExecuteJobArtifacts); optionally scope to a task with taskId. Returns a URL to fetch with your TestMu AI credentials rather than the raw file bytes.
Annotations the server declares
- title: Download Job Artifact
- readOnlyHint: true
- destructiveHint: false
- openWorldHint: false
getAnalyticsDataSummarises web automation test results in a date range: total tests, passed, failed, completed without a pass/fail mark, pass rate (passed over all tests, as the Insights page show…Reads text others writeNo findings
Why it is in this group
- The name "getAnalyticsData" says "get": a word for reading.
What it can reach
- untrusted-inputThe description's first sentence says "summarise" and "web": it reads web pages, which anyone can write.
Description, as the server wrote it Summarises web automation test results in a date range: total tests, passed, failed, completed without a pass/fail mark, pass rate (passed over all tests, as the Insights page shows it), pass rate over tests marked passed or failed, average duration and distinct builds, grouped by status, browser, os, build, device or day. Computed from the automation test list (up to 5000 tests per call; the total is always exact).
Annotations the server declares
- title: Get Analytics Data
- readOnlyHint: true
- destructiveHint: false
- openWorldHint: false
getAnalyticsErrorsSummarises failed and errored web automation tests in a date range: counts by failure category and by status, and the failing tests themselves with build and environment.Reads text others writeNo findings
Why it is in this group
- The name "getAnalyticsErrors" says "get": a word for reading.
What it can reach
- untrusted-inputThe description's first sentence says "summarise" and "web": it reads web pages, which anyone can write.
Description, as the server wrote it Summarises failed and errored web automation tests in a date range: counts by failure category and by status, and the failing tests themselves with build and environment. Failure categories exist only for about the last 7 days of tests; older failures are counted as uncategorised.
Annotations the server declares
- title: Get Analytics Errors
- readOnlyHint: true
- destructiveHint: false
- openWorldHint: false
getAutomationTestBrowserConsoleLogsFetches Browser Console logs for a specific automation test.Reads text others writeNo findings
Why it is in this group
- The name "getAutomationTestBrowserConsoleLogs" says "get": a word for reading.
What it can reach
- untrusted-inputThe name "getAutomationTestBrowserConsoleLogs" says "get" and "browser": it reads web pages, which anyone can write.
- untrusted-inputThe description's first sentence says "fetch" and "browser": it reads web pages, which anyone can write.
Description, as the server wrote it Fetches Browser Console logs for a specific automation test. Note: Not available for mobile app automation tests.
Annotations the server declares
- title: Get Automation Test Browser Console Logs
- readOnlyHint: true
- destructiveHint: false
- openWorldHint: false
getAutomationTestCommandLogsFetches command logs for a specific automation testNo findings
Why it is in this group
- The name "getAutomationTestCommandLogs" says "get": a word for reading.
Description, as the server wrote it Fetches command logs for a specific automation test
Annotations the server declares
- title: Get Automation Test Command Logs
- readOnlyHint: true
- destructiveHint: false
- openWorldHint: false
getAutomationTestDetailsFetches detailed information about an automation test by IDNo findings
Why it is in this group
- The name "getAutomationTestDetails" says "get": a word for reading.
Description, as the server wrote it Fetches detailed information about an automation test by ID
Annotations the server declares
- title: Get Automation Test Details
- readOnlyHint: true
- destructiveHint: false
- openWorldHint: false
getAutomationTestNetworkLogsFetches network logs for a specific automation testNo findings
Why it is in this group
- The name "getAutomationTestNetworkLogs" says "get": a word for reading.
Description, as the server wrote it Fetches network logs for a specific automation test
Annotations the server declares
- title: Get Automation Test Network Logs
- readOnlyHint: true
- destructiveHint: false
- openWorldHint: false
getAutomationTestSeleniumLogsFetches Selenium logs (web), Appium logs (mobile/Appium), or instrumentation logs (mobile/XCUI/Espresso) for a specific automation test.Reads text others writeNo findings
Why it is in this group
- The name "getAutomationTestSeleniumLogs" says "get": a word for reading.
What it can reach
- untrusted-inputThe description's first sentence says "fetch" and "web": it reads web pages, which anyone can write.
Description, as the server wrote it Fetches Selenium logs (web), Appium logs (mobile/Appium), or instrumentation logs (mobile/XCUI/Espresso) for a specific automation test. Automatically detects XCUI and Espresso tests and fetches the correct log type.
Annotations the server declares
- title: Get Automation Test Selenium Logs
- readOnlyHint: true
- openWorldHint: false
- destructiveHint: false
getFlakyTestsLists tests that the insights engine marks as flaky (passing and failing intermittently) in a date range, with each test's flake rate, build and environment.No findings
Why it is in this group
- The name "getFlakyTests" says "get": a word for reading.
Description, as the server wrote it Lists tests that the insights engine marks as flaky (passing and failing intermittently) in a date range, with each test's flake rate, build and environment. Returns only flaky tests, plus how many tests were scanned. Flakiness is only computed for about the last 7 days of tests; the response says how many tests in the range had it.
Annotations the server declares
- title: Get Flaky Tests
- readOnlyHint: true
- destructiveHint: false
- openWorldHint: false
getHyperExecuteJobArtifactsRetrieve all artifacts or categorized errors for a HyperExecute job or task.No findings
Why it is in this group
- The name "getHyperExecuteJobArtifacts" says "get": a word for reading.
Description, as the server wrote it Retrieve all artifacts or categorized errors for a HyperExecute job or task.
Annotations the server declares
- title: Get HyperExecute Job Artifacts
- readOnlyHint: true
- destructiveHint: false
- openWorldHint: false
getHyperExecuteJobInfoFetches and displays information about a HyperExecute job including status, test results, and progress.No findings
Why it is in this group
- The name "getHyperExecuteJobInfo" says "get": a word for reading.
Description, as the server wrote it Fetches and displays information about a HyperExecute job including status, test results, and progress.
Annotations the server declares
- title: Show HyperExecute Job
- readOnlyHint: true
- destructiveHint: false
- openWorldHint: false
getHyperExecuteJobLiveStreamStream live output from a running HyperExecute job.No findings
Why it is in this group
- The name "getHyperExecuteJobLiveStream" says "get": a word for reading.
Description, as the server wrote it Stream live output from a running HyperExecute job. (UNDOCUMENTED - may change without notice)
Annotations the server declares
- title: Get HyperExecute Job Live Stream
- readOnlyHint: true
- destructiveHint: false
- openWorldHint: false
getHyperExecuteJobSessionsFetches and displays session information for a HyperExecute job.No findings
Why it is in this group
- The name "getHyperExecuteJobSessions" says "get": a word for reading.
Description, as the server wrote it Fetches and displays session information for a HyperExecute job.
Annotations the server declares
- title: Show HyperExecute Job Sessions
- readOnlyHint: true
- destructiveHint: false
- openWorldHint: false
getHyperExecuteStageRetrieve the stages of a HyperExecute task (cache/prerun/scenario/etc.).No findings
Why it is in this group
- The name "getHyperExecuteStage" says "get": a word for reading.
Description, as the server wrote it Retrieve the stages of a HyperExecute task (cache/prerun/scenario/etc.). Pass a stageId to return just that one stage. Stages are keyed by their parent task, so taskId is required.
Annotations the server declares
- title: Get HyperExecute Stage
- readOnlyHint: true
- destructiveHint: false
- openWorldHint: false
getHyperExecuteStageLogsRetrieve logs for a HyperExecute stage.No findings
Why it is in this group
- The name "getHyperExecuteStageLogs" says "get": a word for reading.
Description, as the server wrote it Retrieve logs for a HyperExecute stage. (UNDOCUMENTED - may change without notice)
Annotations the server declares
- title: Get HyperExecute Stage Logs
- readOnlyHint: true
- destructiveHint: false
- openWorldHint: false
getHyperExecuteStageTasksRetrieve tasks for a HyperExecute stage.No findings
Why it is in this group
- The name "getHyperExecuteStageTasks" says "get": a word for reading.
Description, as the server wrote it Retrieve tasks for a HyperExecute stage. (UNDOCUMENTED - may change without notice)
Annotations the server declares
- title: Get HyperExecute Stage Tasks
- readOnlyHint: true
- destructiveHint: false
- openWorldHint: false
getHyperExecuteTaskRetrieve details for a HyperExecute task by task ID.No findings
Why it is in this group
- The name "getHyperExecuteTask" says "get": a word for reading.
Description, as the server wrote it Retrieve details for a HyperExecute task by task ID. (UNDOCUMENTED - may change without notice)
Annotations the server declares
- title: Get HyperExecute Task
- readOnlyHint: true
- destructiveHint: false
- openWorldHint: false
getHyperExecuteTaskLogsRetrieve logs for a HyperExecute task.No findings
Why it is in this group
- The name "getHyperExecuteTaskLogs" says "get": a word for reading.
Description, as the server wrote it Retrieve logs for a HyperExecute task. (UNDOCUMENTED - may change without notice)
Annotations the server declares
- title: Get HyperExecute Task Logs
- readOnlyHint: true
- destructiveHint: false
- openWorldHint: false
getJobArtifactDetailGet details for a specific artifact within a HyperExecute job.No findings
Why it is in this group
- The name "getJobArtifactDetail" says "get": a word for reading.
Description, as the server wrote it Get details for a specific artifact within a HyperExecute job.
Annotations the server declares
- title: Get Job Artifact Detail
- readOnlyHint: true
- destructiveHint: false
- openWorldHint: false
getLiveSessionLogsRetrieve the request and response log of a test session, while it runs or after it ends.No findings
Why it is in this group
- The name "getLiveSessionLogs" says "get": a word for reading.
Description, as the server wrote it Retrieve the request and response log of a test session, while it runs or after it ends. Web automation (Selenium, Playwright, Cypress) only: an app automation session returns not found; listMobileSessions carries the video and screenshot links of app sessions.
Annotations the server declares
- title: Get Live Session Logs
- readOnlyHint: true
- destructiveHint: false
- openWorldHint: false
getMobileCapabilityGeneratorRetrieve capability generation data for mobile automation test configuration.No findings
Why it is in this group
- The name "getMobileCapabilityGenerator" says "get": a word for reading.
Description, as the server wrote it Retrieve capability generation data for mobile automation test configuration.
Annotations the server declares
- title: Get Mobile Capability Generator
- readOnlyHint: true
- destructiveHint: false
- openWorldHint: false
getMobileConcurrencyRetrieve the current concurrency usage and limits for mobile automation (real device testing).No findings
Why it is in this group
- The name "getMobileConcurrency" says "get": a word for reading.
Description, as the server wrote it Retrieve the current concurrency usage and limits for mobile automation (real device testing).
Annotations the server declares
- title: Get Mobile Concurrency
- readOnlyHint: true
- destructiveHint: false
- openWorldHint: false
getOrgConcurrencyRetrieve the current concurrency usage and limits for the organization on web automation (Selenium/Cypress).Reads text others writeNo findings
Why it is in this group
- The name "getOrgConcurrency" says "get": a word for reading.
What it can reach
- untrusted-inputThe description's first sentence says "retrieve" and "web": it reads web pages, which anyone can write.
Description, as the server wrote it Retrieve the current concurrency usage and limits for the organization on web automation (Selenium/Cypress).
Annotations the server declares
- title: Get Org Concurrency
- readOnlyHint: true
- destructiveHint: false
- openWorldHint: false
getSessionScreenshotsRetrieve command-by-command screenshots captured during a test session.No findings
Why it is in this group
- The name "getSessionScreenshots" says "get": a word for reading.
Description, as the server wrote it Retrieve command-by-command screenshots captured during a test session. Returns screenshot URLs for each step. Web automation (Selenium, Playwright, Cypress) only: an app automation session returns not found; listMobileSessions carries the video and screenshot links of app sessions.
Annotations the server declares
- title: Get Session Screenshots
- readOnlyHint: true
- destructiveHint: false
- openWorldHint: false
getSessionV2CommandLogsRetrieve v2 command execution logs for a test session.No findings
Why it is in this group
- The name "getSessionV2CommandLogs" says "get": a word for reading.
Description, as the server wrote it Retrieve v2 command execution logs for a test session. Returns structured JSON with enhanced detail compared to v1 logs. Web automation (Selenium, Playwright, Cypress) only: an app automation session returns not found; listMobileSessions carries the video and screenshot links of app sessions.
Annotations the server declares
- title: Get Session V2 Command Logs
- readOnlyHint: true
- destructiveHint: false
- openWorldHint: false
getSessionV2SeleniumLogsRetrieve v2 Selenium/WebDriver grid request and response logs for a test session.Reads text others writeNo findings
Why it is in this group
- The name "getSessionV2SeleniumLogs" says "get": a word for reading.
What it can reach
- untrusted-inputThe description's first sentence says "retrieve" and "web": it reads web pages, which anyone can write.
Description, as the server wrote it Retrieve v2 Selenium/WebDriver grid request and response logs for a test session. Web automation (Selenium, Playwright, Cypress) only: an app automation session returns not found; listMobileSessions carries the video and screenshot links of app sessions.
Annotations the server declares
- title: Get Session V2 Selenium Logs
- readOnlyHint: true
- destructiveHint: false
- openWorldHint: false
getSessionVideoRetrieve the video recording URL for a completed test session.Reads text others writeNo findings
Why it is in this group
- The name "getSessionVideo" says "get": a word for reading.
What it can reach
- untrusted-inputThe description's first sentence says "retrieve" and "url": it reads web pages, which anyone can write.
Description, as the server wrote it Retrieve the video recording URL for a completed test session. Web automation (Selenium, Playwright, Cypress) only: an app automation session returns not found; listMobileSessions carries the video and screenshot links of app sessions.
Annotations the server declares
- title: Get Session Video
- readOnlyHint: true
- destructiveHint: false
- openWorldHint: false
getSuiteOverviewGets comprehensive suite overview for a project including run counts, latest status, scores, schedule information, and active calls.No findings
Why it is in this group
- The name "getSuiteOverview" says "get": a word for reading.
Description, as the server wrote it Gets comprehensive suite overview for a project including run counts, latest status, scores, schedule information, and active calls.
Annotations the server declares
- title: Get Suite Overview
- readOnlyHint: true
- destructiveHint: false
- openWorldHint: false
getSupportedPlatformsLists the OS, browser and device combinations TestMu AI supports, to find valid os, browser and version values for a test configuration.Reads text others writeNo findings
Why it is in this group
- The name "getSupportedPlatforms" says "get": a word for reading.
What it can reach
- untrusted-inputThe description's first sentence says "list" and "browser": it reads web pages, which anyone can write.
Description, as the server wrote it Lists the OS, browser and device combinations TestMu AI supports, to find valid os, browser and version values for a test configuration. With no filters it returns a compact summary: each desktop OS with its browsers (version count, newest and oldest) and resolutions, and each mobile platform with its device count and OS versions. Set os (and optionally browser) for the full version list, or os=ios / os=android for device names.
Annotations the server declares
- title: Get Supported Platforms
- readOnlyHint: true
- destructiveHint: false
- openWorldHint: false
getTestTrendsDaily trend of web automation test results in a date range: for each day, total tests, passed, failed, completed without a pass/fail mark, pass rate (passed over all tests), pass r…Reads text others writeNo findings
Why it is in this group
- The name "getTestTrends" says "get": a word for reading.
What it can reach
- untrusted-inputThe description's first sentence says "web": it reads web pages, which anyone can write.
Description, as the server wrote it Daily trend of web automation test results in a date range: for each day, total tests, passed, failed, completed without a pass/fail mark, pass rate (passed over all tests), pass rate over tests marked passed or failed, and average duration.
Annotations the server declares
- title: Get Test Trends
- readOnlyHint: true
- destructiveHint: false
- openWorldHint: false
listDevicesForTestingLists the real devices available on TestMu AI for mobile app testing, 50 per page by default.No findings
Why it is in this group
- The name "listDevicesForTesting" says "list": a word for reading.
Description, as the server wrote it Lists the real devices available on TestMu AI for mobile app testing, 50 per page by default. Filter by region, OS, device name or OS version, and page with limit and offset. The text gives the total that matched.
Annotations the server declares
- title: List Devices for Testing
- readOnlyHint: true
- destructiveHint: false
- openWorldHint: false
listHyperExecuteJobsList the account's HyperExecute jobs, newest first, with offset pagination and an optional status filter.No findings
Why it is in this group
- The name "listHyperExecuteJobs" says "list": a word for reading.
Description, as the server wrote it List the account's HyperExecute jobs, newest first, with offset pagination and an optional status filter. The status filter looks at the 200 most recent jobs.
Annotations the server declares
- title: List HyperExecute Jobs
- readOnlyHint: true
- destructiveHint: false
- openWorldHint: false
listMobileBuildsLists mobile app automation builds, newest first, 10 per page by default.No findings
Why it is in this group
- The name "listMobileBuilds" says "list": a word for reading.
Description, as the server wrote it Lists mobile app automation builds, newest first, 10 per page by default. Page with limit and offset; the result gives the total the API reports. The API returns only builds from the last 90 days, so total can be higher than the builds you can page through.
Annotations the server declares
- title: List Mobile Builds
- readOnlyHint: true
- destructiveHint: false
- openWorldHint: false
listMobileSessionsLists mobile app automation sessions (tests), newest first, 10 per page by default, with their video, screenshot and log links.No findings
Why it is in this group
- The name "listMobileSessions" says "list": a word for reading.
Description, as the server wrote it Lists mobile app automation sessions (tests), newest first, 10 per page by default, with their video, screenshot and log links. Page with limit and offset, or filter by buildId or status; the result gives the total the API reports.
Annotations the server declares
- title: List Mobile Sessions
- readOnlyHint: true
- destructiveHint: false
- openWorldHint: false
listTunnelsRetrieve all active TestMu AI tunnel connections for the account.No findings
Why it is in this group
- The name "listTunnels" says "list": a word for reading.
Description, as the server wrote it Retrieve all active TestMu AI tunnel connections for the account. Useful for checking which tunnels are running before starting a new one or debugging local testing connectivity.
Annotations the server declares
- title: List Tunnels
- readOnlyHint: true
- destructiveHint: false
- openWorldHint: false
listUploadedAppsRetrieve all apps uploaded to TestMu AI for mobile testing.No findings
Why it is in this group
- The name "listUploadedApps" says "list": a word for reading.
Description, as the server wrote it Retrieve all apps uploaded to TestMu AI for mobile testing. Use type=android/ios for real devices, emulator/simulator for virtual devices.
Annotations the server declares
- title: List Uploaded Apps
- readOnlyHint: true
- destructiveHint: false
- openWorldHint: false
smartui_build_queryList SmartUI builds in a project, get one build, or get the CI gate verdict for a build.No findings
Why it is in this group
- The name "smartui_build_query" says "query": a word for reading.
Description, as the server wrote it List SmartUI builds in a project, get one build, or get the CI gate verdict for a build. Use action=gate in pipelines: it returns verdict (passed|failed|pending_approval|running|error), totals (including missing screenshots the baseline has), blocking_comparison_ids, and blocking[] with per-item detail. verdict=failed covers rejections AND builds that dropped baseline screenshots. Read-only.
Annotations the server declares
- title: SmartUI Build Query
- readOnlyHint: true
- destructiveHint: false
- openWorldHint: false
smartui_comparison_analyzeAnalyze one SmartUI comparison through a lens: consolidated (default), pixel, or dom.No findings
Why it is in this group
- The name "smartui_comparison_analyze" says "analyze": a word for reading.
Description, as the server wrote it Analyze one SmartUI comparison through a lens: consolidated (default), pixel, or dom. Returns the comparison data (mismatch %, verdict, correctly-labelled asset URLs) plus, for lens=dom, a real DOM byte delta when the artefact is served — no local filesystem needed. consolidated and pixel report upstream numbers and point at the assets; they do not run their own image analysis. lens=dom degrades to a structured error when DOM artefacts are not served. Read-only.
Annotations the server declares
- title: SmartUI Comparison Analyze
- readOnlyHint: true
- destructiveHint: false
- openWorldHint: false
smartui_comparison_queryList all comparisons in a SmartUI build (each row returns comparison_id AND ss_id), or get one comparison's detail (mismatch %, status, correctly-labelled baseline/current/compared…No findings
Why it is in this group
- The name "smartui_comparison_query" says "query": a word for reading.
Description, as the server wrote it List all comparisons in a SmartUI build (each row returns comparison_id AND ss_id), or get one comparison's detail (mismatch %, status, correctly-labelled baseline/current/compared asset URLs). This is how you obtain a comparison_id — you never need the dashboard. Read-only.
Annotations the server declares
- title: SmartUI Comparison Query
- readOnlyHint: true
- destructiveHint: false
- openWorldHint: false
smartui_project_queryList SmartUI projects, or get one project with its settings — action=get returns baseline_branch, approvers[], smart_baseline, auto_approval_branch, overwrite_screenshot and projec…Reads private dataNo findings
Why it is in this group
- The name "smartui_project_query" says "query": a word for reading.
What it can reach
- private-dataThe description's first sentence says "list" and "branch": it reads repositories.
Description, as the server wrote it List SmartUI projects, or get one project with its settings — action=get returns baseline_branch, approvers[], smart_baseline, auto_approval_branch, overwrite_screenshot and project_category alongside platform and project_type. Start here when you do not have a project_id. Read-only.
Annotations the server declares
- title: SmartUI Project Query
- readOnlyHint: true
- destructiveHint: false
- openWorldHint: false
tm_get_coverage_summaryTest coverage for a project, overall and per folder, computed live from the project's test cases (matches the Test Cases screen).Reads private dataNo findings
Why it is in this group
- The name "tm_get_coverage_summary" says "get": a word for reading.
What it can reach
- private-dataThe description's first sentence says "folder": it reads files.
- private-dataThe input names a folder (input.folder_id), and the tool reads: it reads files.
Description, as the server wrote it Test coverage for a project, overall and per folder, computed live from the project's test cases (matches the Test Cases screen). Also answers requirement and traceability coverage: pass jira_id (any Jira issue key, e.g. TE-123: a requirement, user story, epic, ticket or bug) to get that item's test coverage (the test cases linked to it and how many are automated; zero linked cases means uncovered). For a set of requirements, call once per Jira key. A scope with no test cases returns all zeros; a Jira key that does not exist in the connected Jira returns an error.
Annotations the server declares
- title: Get Coverage Summary
- readOnlyHint: true
- destructiveHint: false
- openWorldHint: false
tm_get_milestonePass milestone_id to get one milestone's detail and live completion progress.No findings
Why it is in this group
- The name "tm_get_milestone" says "get": a word for reading.
Description, as the server wrote it Pass milestone_id to get one milestone's detail and live completion progress. Otherwise pass project_id to list the project's milestones.
Annotations the server declares
- title: Get Milestones
- readOnlyHint: true
- destructiveHint: false
- openWorldHint: false
tm_get_test_casesFind test cases in a project by filters (folder, priority, automation status, type, tags) and/or search (matches test case ID like TC-123, or title; minimum 3 characters) — returns…Reads private dataNo findings
Why it is in this group
- The name "tm_get_test_cases" says "get": a word for reading.
What it can reach
- private-dataThe description's first sentence says "find" and "folder": it reads files.
- private-dataThe input names a folder (input.folder_id), and the tool reads: it reads files.
Description, as the server wrote it Find test cases in a project by filters (folder, priority, automation status, type, tags) and/or search (matches test case ID like TC-123, or title; minimum 3 characters) — returns summaries. Filters combine (AND). Pass case_id instead to get one case's full detail (steps, BDD scenarios, field values).
Annotations the server declares
- title: Find / Get Test Cases
- readOnlyHint: true
- destructiveHint: false
- openWorldHint: false
tm_get_test_runsPass run_id to get one test run's detail plus its result summary (instance counts by status — passed / failed / skipped / not started / in progress — and pass rate).Reads private dataNo findings
Why it is in this group
- The name "tm_get_test_runs" says "get": a word for reading.
What it can reach
- private-dataThe input names a folder (input.folder_id), and the tool reads: it reads files.
Description, as the server wrote it Pass run_id to get one test run's detail plus its result summary (instance counts by status — passed / failed / skipped / not started / in progress — and pass rate). Otherwise pass project_id to list runs with optional filters. Every returned run carries run_type ("kaneai" or "manual") identifying KaneAI runs — the backend stores KaneAI runs with type "Manual" (result-entry mode), so use run_type, not type, to tell them apart.
Annotations the server declares
- title: Get Test Runs
- readOnlyHint: true
- destructiveHint: false
- openWorldHint: false
tm_list_configurationsList saved run Configurations (reusable platform / browser / OS / device setups).Reads text others writeNo findings
Why it is in this group
- The name "tm_list_configurations" says "list": a word for reading.
What it can reach
- untrusted-inputThe description's first sentence says "list" and "browser": it reads web pages, which anyone can write.
Description, as the server wrote it List saved run Configurations (reusable platform / browser / OS / device setups). To apply one to a run, pass the configuration's TOP-LEVEL id to tm_create_test_run or tm_update_test_run as configuration_id (NOT the nested environments[].environment_id). Pass run_type to see only the configurations usable for that run type (kaneai vs manual). Each test instance carries exactly one configuration. Configurations are created by humans in the Test Manager UI. They are organization-level (shared by every project), so there is no project filter. Results are paged: the response says how many exist in total; page through them or raise limit to see all.
Annotations the server declares
- title: List Run Configurations
- readOnlyHint: true
- destructiveHint: false
- openWorldHint: false
tm_list_foldersGet a folder tree.Reads private dataNo findings
Why it is in this group
- The name "tm_list_folders" says "list": a word for reading.
What it can reach
- private-dataThe name "tm_list_folders" says "list" and "folder": it reads files.
- private-dataThe description's first sentence says "get" and "folder": it reads files.
Description, as the server wrote it Get a folder tree. Test Manager keeps three folder types: Test Cases and Test Runs folders (per project — pass project_id) and Modules folders (organization-level, since modules are shared across projects — no project_id needed). Pick with the tree parameter (default: test-cases). New test-case/test-run folders can be created with tm_create_folder.
Annotations the server declares
- title: List Folders
- readOnlyHint: true
- destructiveHint: false
- openWorldHint: false
tm_list_modulesList the organization's reusable step Modules (modules are org-level, shared across projects — the same list as the Modules page).Reads private dataNo findings
Why it is in this group
- The name "tm_list_modules" says "list": a word for reading.
What it can reach
- private-dataThe input names a folder (input.folder_id), and the tool reads: it reads files.
Description, as the server wrote it List the organization's reusable step Modules (modules are org-level, shared across projects — the same list as the Modules page). Supports name search and pagination. Reference a module by id when creating or editing a test case to include its steps (the Module itself is never changed).
Annotations the server declares
- title: List Step Modules
- readOnlyHint: true
- destructiveHint: false
- openWorldHint: false
tm_list_projectsList all Test Manager projects the authenticated user is allowed to see.No findings
Why it is in this group
- The name "tm_list_projects" says "list": a word for reading.
Description, as the server wrote it List all Test Manager projects the authenticated user is allowed to see. Supports keyword search and pagination.
Annotations the server declares
- title: List Test Manager Projects
- readOnlyHint: true
- destructiveHint: false
- openWorldHint: false
This is not a safety rating, and no finding is not a safety guarantee. The groups and findings come from fixed checks of the tool definitions the server returned on the date shown, not from its code or what it does when called. How the checks work.
Exfil surface
Which tools read private data, which read text someone else can write, and which can send data out. One of each is a path a prompt injection can use.
560 paths in this server alone from private data, through text someone else wrote, to a way out. Denying 5 outbound tools breaks every one.
- Reads private data7
- Reads files, mail, messages, databases, secrets, calendars or repositories.
- Reads text others write16
- Brings in text someone else can write: web pages, inbound email, messages, issues and comments.
- Can send data out5
- Can carry data to people or systems: it sends, posts or publishes, or takes a recipient, channel or URL.
{
"permissions": {
"deny": [
"mcp__lambdatest__getAccessibilityReport",
"mcp__lambdatest__postSessionExceptions",
"mcp__lambdatest__postTestExceptions",
"mcp__lambdatest__tm_link_ado_issue",
"mcp__lambdatest__tm_link_jira_issue"
]
}
}Other servers in the same agent add their own tools to these paths. Audit your whole setup to see them together.
Remote posture
What anyone on the internet sees at the remote endpoint: an anonymous initialize and tools/list, and the public OAuth metadata.
Looked at Nov 20, 2026, 18:59 UTC, unchanged since Oct 29, 2026. No credentials were sent and no tool was called. What is looked at.
- Reachable
- yes
- HTTP status of an anonymous initialize
- 200
- tools/list served without credentials
- yes
- Tools that change things, listed without credentials
- 37
- Protocol version
- 2025-03-26
- WWW-Authenticate on 401
- not observed
- resource_metadata in WWW-Authenticate
- not observed
- RFC 9728 protected resource metadata
- https://mcp.lambdatest.com/.well-known/oauth-protected-resource/mcp
- Authorization servers
- https://auth.lambdatest.com
- Legacy HTTP+SSE transport
- no
- Dynamic client registration
- yes
- Client ID metadata documents
- no
- HighThe whole server
The endpoint served tools/list without credentials, and the list has 37 tools that change things (19 destructive, 18 write). Whether a call also works without credentials was not tested: tools are never called.
What this check looks forMatched text
https://mcp.lambdatest.com/mcp: tools/list returned 88 tools to an anonymous client, including updateBuild (destructive), stopBuild (destructive), deleteBuild (destructive), stopTunnel (destructive), deleteSession (destructive) and 32 more
Posture changes
- Oct 29, 2026
- Tools that change things, listed without credentials: 36 to 37
Findings
Matches of fixed checks, with the text they matched.
The server marks "generateHyperExecuteYAML" read-only (readOnlyHint: true), and its own name, description or input give the class "write".
What this check looks forMatched text
annotations: {"readOnlyHint":true,"destructiveHint":false,"openWorldHint":false}; The name "generateHyperExecuteYAML" says "generate": a word for a change.
The server marks "answerHyperExecuteQuery" read-only (readOnlyHint: true), and its own name, description or input give the class "write".
What this check looks forMatched text
annotations: {"readOnlyHint":true,"destructiveHint":false,"openWorldHint":false}; The name "answerHyperExecuteQuery" says "execute": a word for a change.
The server marks "analyzeAppViaTunnel" as not read-only (readOnlyHint: false), and its words say it reads.
What this check looks forMatched text
annotations: {"readOnlyHint":false,"destructiveHint":false,"openWorldHint":false}; The name "analyzeAppViaTunnel" says "analyze": a word for reading.
The server marks "checkAppProcessingStatus" as not read-only (readOnlyHint: false), and its words say it reads.
What this check looks forMatched text
annotations: {"readOnlyHint":false,"destructiveHint":false,"idempotentHint":true,"openWorldHint":false}; The name "checkAppProcessingStatus" says "status": a word for reading.
The server marks "smartui_screenshot_upload" destructive (destructiveHint: true), and its name says "screenshot", a word for reading.
What this check looks forMatched text
annotations: {"readOnlyHint":false,"destructiveHint":true,"openWorldHint":false}; The name "smartui_screenshot_upload" says "screenshot": a word for reading.
The server marks "smartui_screenshot_review" destructive (destructiveHint: true), and its name says "screenshot", a word for reading.
What this check looks forMatched text
annotations: {"readOnlyHint":false,"destructiveHint":true,"openWorldHint":false}; The name "smartui_screenshot_review" says "screenshot": a word for reading.
Show 1 more findingShow fewer
The description mentions "credentials".
What this check looks forMatched text
description: "...urns a URL to fetch with your TestMu AI credentials rather than the raw file bytes."
History
Changes to its tool definitions and to its remote endpoint, newest first.
Tools20 changed.~20 changedView diff
- Changed:analyzeAppViaTunnelChanges thingsdescriptionannotations
- Changed:automationQueryOnly readsdescription
- Changed:buildLocalAppForAnalysisOnly readsdescription
- Changed:checkAppProcessingStatusChanges thingsdescription
- Changed:deleteBuildDeletes or cancelsdescription
- Changed:deletePrerunFileDeletes or cancelsdescription
- Changed:deleteSessionDeletes or cancelsdescription
- Changed:deleteUploadedAppsDeletes or cancelsdescription
- Changed:deleteUserFileDeletes or cancelsdescription
- Changed:downloadJobArtifactOnly readsdescription
- Changed:generateHyperExecuteYAMLChanges thingsdescription
- Changed:getSupportedPlatformsOnly readsdescription
- Changed:listDevicesForTestingOnly readsdescription
- Changed:listTunnelsOnly readsdescription
- Changed:listUploadedAppsOnly readsdescription
- Changed:stopBuildDeletes or cancelsdescription
- Changed:stopSessionDeletes or cancelsdescription
- Changed:updateBuildDeletes or cancelsdescription
- Changed:uploadAppVirtualDeviceChanges thingsdescription
- Changed:upload_appChanges thingsdescription
PostureRemote posture changed. Tools that change things, listed without credentials: 36 -> 37.View diff
- Tools that change things, listed without credentials: 36 to 37
Install
Pinned to version 1.0.7. Values in angle brackets are yours to fill in.
claude mcp add --transport http lambdatest https://mcp.lambdatest.com/mcpclaude mcp add --transport http lambdatest https://mcp.lambdatest.com/mcpRun in a terminal.
Claude Code permissions
Allow the reads, ask before a change or a way out, deny the deletes. Paste into .claude/settings.json.
{
"permissions": {
"allow": [
"mcp__lambdatest__getHyperExecuteJobInfo",
"mcp__lambdatest__getHyperExecuteJobSessions",
"mcp__lambdatest__buildLocalAppForAnalysis",
"mcp__lambdatest__getAutomationTestDetails",
"mcp__lambdatest__getAutomationTestCommandLogs",
"mcp__lambdatest__getAutomationTestNetworkLogs",
"mcp__lambdatest__getAutomationTestBrowserConsoleLogs",
"mcp__lambdatest__getAutomationTestSeleniumLogs",
"mcp__lambdatest__automationQuery",
"mcp__lambdatest__getSuiteOverview",
"mcp__lambdatest__getSupportedPlatforms",
"mcp__lambdatest__listTunnels",
"mcp__lambdatest__getSessionScreenshots",
"mcp__lambdatest__getSessionVideo",
"mcp__lambdatest__getSessionV2CommandLogs",
"mcp__lambdatest__getSessionV2SeleniumLogs",
"mcp__lambdatest__getLiveSessionLogs",
"mcp__lambdatest__getOrgConcurrency",
"mcp__lambdatest__getAnalyticsData",
"mcp__lambdatest__getFlakyTests",
"mcp__lambdatest__getTestTrends",
"mcp__lambdatest__getAnalyticsErrors",
"mcp__lambdatest__listUploadedApps",
"mcp__lambdatest__listDevicesForTesting",
"mcp__lambdatest__getMobileCapabilityGenerator",
"mcp__lambdatest__listMobileBuilds",
"mcp__lambdatest__listMobileSessions",
"mcp__lambdatest__getMobileConcurrency",
"mcp__lambdatest__tm_list_projects",
"mcp__lambdatest__tm_get_test_cases",
"mcp__lambdatest__tm_list_folders",
"mcp__lambdatest__tm_list_modules",
"mcp__lambdatest__tm_list_configurations",
"mcp__lambdatest__tm_get_test_runs",
"mcp__lambdatest__tm_get_milestone",
"mcp__lambdatest__tm_get_coverage_summary",
"mcp__lambdatest__smartui_project_query",
"mcp__lambdatest__smartui_build_query",
"mcp__lambdatest__smartui_comparison_query",
"mcp__lambdatest__smartui_comparison_analyze",
"mcp__lambdatest__listHyperExecuteJobs",
"mcp__lambdatest__getHyperExecuteJobArtifacts",
"mcp__lambdatest__downloadJobArtifact",
"mcp__lambdatest__getJobArtifactDetail",
"mcp__lambdatest__getHyperExecuteStage",
"mcp__lambdatest__getHyperExecuteStageLogs",
"mcp__lambdatest__getHyperExecuteStageTasks",
"mcp__lambdatest__getHyperExecuteTask",
"mcp__lambdatest__getHyperExecuteTaskLogs",
"mcp__lambdatest__getHyperExecuteJobLiveStream",
"mcp__lambdatest__analyzeHyperExecuteJob"
],
"ask": [
"mcp__lambdatest__upload_app",
"mcp__lambdatest__generateHyperExecuteYAML",
"mcp__lambdatest__answerHyperExecuteQuery",
"mcp__lambdatest__getAccessibilityReport",
"mcp__lambdatest__analyzeAppViaTunnel",
"mcp__lambdatest__triggerSuiteRun",
"mcp__lambdatest__postSessionExceptions",
"mcp__lambdatest__postTestExceptions",
"mcp__lambdatest__uploadAppVirtualDevice",
"mcp__lambdatest__checkAppProcessingStatus",
"mcp__lambdatest__tm_create_project",
"mcp__lambdatest__tm_create_folder",
"mcp__lambdatest__tm_generate_test_cases",
"mcp__lambdatest__tm_create_test_case",
"mcp__lambdatest__tm_create_test_run",
"mcp__lambdatest__tm_create_milestone",
"mcp__lambdatest__tm_link_jira_issue",
"mcp__lambdatest__tm_link_ado_issue"
],
"deny": [
"mcp__lambdatest__updateBuild",
"mcp__lambdatest__stopBuild",
"mcp__lambdatest__deleteBuild",
"mcp__lambdatest__stopTunnel",
"mcp__lambdatest__deleteSession",
"mcp__lambdatest__updateSession",
"mcp__lambdatest__stopSession",
"mcp__lambdatest__deletePrerunFile",
"mcp__lambdatest__deleteUserFile",
"mcp__lambdatest__deleteUploadedApps",
"mcp__lambdatest__tm_update_project",
"mcp__lambdatest__tm_update_folder",
"mcp__lambdatest__tm_update_test_case",
"mcp__lambdatest__tm_update_test_run",
"mcp__lambdatest__tm_record_test_results",
"mcp__lambdatest__tm_update_milestone",
"mcp__lambdatest__smartui_screenshot_upload",
"mcp__lambdatest__smartui_screenshot_review",
"mcp__lambdatest__smartui_baseline_manage"
]
}
}Tools that only read are allowed, tools that change things ask first, tools that delete or cancel are denied. A read that can send data out (it takes a URL, say) asks first too. It assumes the server is installed as lambdatest. Review it before you use it.
Capability tags behind 26 rules
- mcp__lambdatest__getAccessibilityReport untrusted-input, outbound
- mcp__lambdatest__buildLocalAppForAnalysis untrusted-input
- mcp__lambdatest__getAutomationTestBrowserConsoleLogs untrusted-input
- mcp__lambdatest__getAutomationTestSeleniumLogs untrusted-input
- mcp__lambdatest__triggerSuiteRun untrusted-input
- mcp__lambdatest__getSupportedPlatforms untrusted-input
- mcp__lambdatest__getSessionVideo untrusted-input
- mcp__lambdatest__getSessionV2SeleniumLogs untrusted-input
- mcp__lambdatest__postSessionExceptions outbound
- mcp__lambdatest__postTestExceptions outbound
- mcp__lambdatest__getOrgConcurrency untrusted-input
- mcp__lambdatest__getAnalyticsData untrusted-input
- mcp__lambdatest__getTestTrends untrusted-input
- mcp__lambdatest__getAnalyticsErrors untrusted-input
- mcp__lambdatest__tm_get_test_cases private-data
- mcp__lambdatest__tm_list_folders private-data
- mcp__lambdatest__tm_list_modules private-data
- mcp__lambdatest__tm_list_configurations untrusted-input
- mcp__lambdatest__tm_get_test_runs private-data
- mcp__lambdatest__tm_get_coverage_summary private-data
- mcp__lambdatest__tm_link_jira_issue outbound
- mcp__lambdatest__tm_link_ado_issue outbound
- mcp__lambdatest__smartui_project_query private-data
- mcp__lambdatest__smartui_screenshot_upload private-data, untrusted-input
- mcp__lambdatest__smartui_screenshot_review untrusted-input
- mcp__lambdatest__downloadJobArtifact untrusted-input
Pin
Keep the decision: fail CI when the tools change, follow every change, or show a badge.
Pin in CI
Fail a build when the tools you reviewed are not the tools the server serves today.
curl -s "https://radar.anlyon.com/api/v1/servers/io.github.LambdaTest/mcp/verify?hash=1b8bf1147ac32b5fdf2a3c8e2a4e934fb283887a9eecd2db49f48d3be735b363" | jq -e '.changed == false'Pinned hash 1b8bf1147ac3, from Nov 16, 2026, 00:17 UTC.
Watch
Get told when its tools change: a new tool, a new description, a new class.
https://radar.anlyon.com/servers/io.github.LambdaTest/mcp/changes.xmlAn Atom feed. Works in any feed reader or Slack's RSS app.
Badge
[](https://radar.anlyon.com/servers/io.github.LambdaTest/mcp)Source: MCP Registry · repository. Synced Oct 21, 2026, 18:17 UTC.
TestMu AI (formerly LambdaTest) MCP server: questions and answers
Answered from the scan above, not from the publisher's description.
Is the TestMu AI (formerly LambdaTest) MCP server safe?
- No scan can promise that. What its definitions show: 88 tools, 37 of which can change, send or delete something, and 7 findings from fixed checks. Allow the reads, ask before each change, and deny the deletes; the permissions block on this page does that.
Can the TestMu AI (formerly LambdaTest) MCP server delete data?
- Yes. 19 tools can delete, cancel or revoke something: updateBuild, stopBuild, deleteBuild, stopTunnel, deleteSession, updateSession and 13 more.
Can the TestMu AI (formerly LambdaTest) MCP server do things that cannot be undone?
- No tool is classed as irreversible: none of them sends, pays, publishes or merges, going by their definitions.
What tools does the TestMu AI (formerly LambdaTest) MCP server have?
- 88 tools: 19 destructive, 18 write, 51 read. updateBuild, stopBuild, deleteBuild, stopTunnel, deleteSession, updateSession, stopSession, deletePrerunFile and 80 more.
Does the TestMu AI (formerly LambdaTest) MCP server need sign-in or OAuth?
- Its remote endpoint listed its tools to an anonymous client, without sign-in.
Does the TestMu AI (formerly LambdaTest) MCP server send data to outside services?
- 5 tools can carry data out (it sends, posts or publishes, or takes a recipient, channel or URL): getAccessibilityReport, postSessionExceptions, postTestExceptions, tm_link_jira_issue and tm_link_ado_issue.
Has the TestMu AI (formerly LambdaTest) MCP server changed its tools?
- Yes, once since Anlyon Radar started tracking it, most recently on Nov 2, 2026. Every change is on this page and in its Atom feed.